Capturing Rapid Vehicle Dynamics: A Refined Adaptive Kalman Filter for Enhanced Tire–Road Friction Estimation

This article proposes an adaptive strong tracking Kalman filter (STKF) to enhance the estimation performance of the tire-road friction coefficient (TRFC), a critical factor in vehicle control and safety. First, a vehicle dynamics model based on a quasi-steady-state tire model is introduced to improve the accuracy of modeling the vehicle response to continuous acceleration fluctuations. Second, an adaptive STKF is developed using the vehicle dynamics model, incorporating a fading factor and an excitation factor. The fading factor, determined from the residual sequence of the observation variable, is utilized to correct the prediction error variance, thereby enhancing the tracking capability and estimation accuracy of the algorithm. The excitation factor, calculated based on real-time vehicle state, adjusts the fading factor to prevent significant prediction error variance under low excitation intensity and to avoid divergence of the estimation algorithm. Experimental results under various road conditions demonstrate the effectiveness of the proposed method in improving the convergence speed and stability of TRFC estimation. The root-mean-square error of the estimated values is improved by more than 18%, and the mean absolute error is improved by over 34% compared to the benchmark method.